The relative importance of information items and preferred mode of delivery when disseminating results from trials to participants: A mixed‐methods study

Abstract Background Participants want to receive the results of trials that they have participated in. Dissemination practices are disparate, and there is limited guidance available on what information to provide to participants and how to deliver it. Objectives This study aimed to establish what trial participants believe should be included in a results summary and how this information should be delivered. Methods A mixed‐methods design was used with focus groups and interviews involving women convenience‐sampled from two host randomized‐controlled trials. Participants ranked information items in order of their importance for inclusion in a trial results summary and potential modes of delivery by preference. All participants provided written informed consent. Results Sixteen women (mean age [SD] = 71.6 [9.7] years) participated. Participants ranked ‘individual results from the study’ and ‘summary of overall trial results’ as most important. Themes such as reassurance and setting results in context were identified as contributing to participants' decisions around ranking. ‘A thank you for your contribution to the study’ was ranked the least important. Delivery by post was the preferred mode of receiving results, with receiving a hard copy of results cited as helpful to refer back to. Conclusion Our findings provide insight into what information trial participants deem as important when receiving trial results and how they would like results delivered. Involving patients during development of trial results to be communicated to participants could help to ensure that the right information is delivered in the right way. Patient or Public Contribution Public partners were involved in focussed aspects of study conduct.


| BACKGROUND
The importance of disseminating results to people participating in clinical trials has been highlighted in recent years based on participants' requests to receive results, which aligns with efforts to increase transparency in research. [1][2][3] Current best practice guidance on an optimal approach to trial results dissemination to participants is limited.
A recent study found that the majority of trials (87.7%) intended to report findings to participants. 4 However, only 18.8% reported an active effort to disseminate results, with the majority of trial teams (80.5%) placing the responsibility on local clinical care teams or leaving the onus on participants themselves to actively locate the results. 4 A further investigation examining dissemination outcomes in 1818 studies found that only 40% had disseminated results to trial participants (or planned to) up to 2 years after publication. 5 Various barriers have been reported as to why researchers do not widely disseminate results and include the following: their own perceptions that participants would not be interested in receiving the results and lack of knowledge of how to disseminate results to lay audiences. 2,5 The Health Research Authority (HRA), who are a unified national system for health research governance in the United Kingdom, published guidance in 2015 on the type of information to provide study participants at the end of study and suggest that this is in the form of a summary form, written in lay language. 6,7 Although there is guidance in place, there remains a lack of clarity in determining the content and mode of delivery for trial results.
Previous research on trial dissemination is limited, being undertaken largely in one clinical context (oncology-based trials) or based on hypothetical trial scenarios presented to participants. [8][9][10][11][12][13][14][15] Additionally, the focus of much of the research to date has been participant comprehension of results rather than exploring what information participants wanted to know or how to deliver this information to them. 10,16,17 Understanding which information trial participants deem to be important and how best they would like trial results to be disseminated may provide valuable insight for trial teams to better prepare their dissemination plans. The aim of our study was to de-termine what information participants with lived experience of trial participation believe should be included in a summary of trial results and how this information should be delivered to them.

| Study design
The study design was an explanatory mixed-methods design, in which the data were collected concurrently, but analysed sequentially (i.e., quantitative data analysed first and qualitative data analysed to inform the quantitative findings). This mixed-methods design was used to allow a more in-depth understanding of the reasons why women valued certain information items or modes of delivery over others but also to allow more general perspectives about the provision of trial result summaries to be explored. 18

| Participants and consent
Potential participants were convenience-sampled from two host trials involving female participants (PROSPECT, which had two parallel randomized trials evaluating different surgeries for primary transvaginal anterior or posterior compartment prolapse surgery [ISRCTN60695184], 19  and women could opt into taking part in the study by returning a reply slip or contacting the study team. Reminders were sent approximately 2 weeks following initial invitations. All participants provided written informed consent.

| Preparation of resource materials
Existing literature and the HRA guidance 6 on feedback of trial results to participants were used to inform the topic guides (Appendix 1) for the focus groups and interviews. The topic guide was semi-structured in format and focused on areas relevant to the study's overarching aims, but was flexible to allow participants to express their opinions and views. The topic guide and prompts were further developed and refined through piloting with colleagues and the research team before the planned focus group discussions. Two ranking exercises to help participants prioritize the information and mode were developed based on the suggested information items and modes of delivery in the HRA guidance. 6 The information items included information content items (contact details for questions, how to report side effects, information on what happens after the study, invitation for future research, study title and publication reference, summary of individual results, summary of results, summary of study, thank you for participation, treatment received) and items on mode of delivery (DVD, email, in-person, link to website, post). In advance of participants ranking the information items, they were asked to consider what information they would want to know at the end of a trial.
Focus groups and interviews were chosen to provide more in-depth investigation of information preferences generated in combination with the ranking exercises. The use of focus groups also enabled participants to hear and consider other people's views and preferences.

| Focus groups and interviews
Focus groups with PROSPECT participants were conducted in March 2018, and interviews with VUE participants were conducted between November and December 2019 at the Heath Services Research Unit, University of Aberdeen. Focus groups were the research team's preferred mode of data collection for this project. However, to maximize participation, from the outset, participants were offered the option to attend an interview if they could not attend a focus group.
Due to participant availability, interviews were conducted with participants from the second trial cohort (VUE trial). Focus groups were moderated by the study team (Authors 2 and 3) and interviews were conducted by the lead authors. All participants completed both ranking exercises; in PROSPECT, this was done as an electronic group ranking activity, which allowed an immediate summary of results.
Items were ranked from 1 to 10 (1 = most important, 10 = least important) or 1 to 5 (1 = most preferred, 5 = least preferred) and also completed individually in the VUE interviews and completed as an electronic group ranking activity in the PROSPECT focus groups.
Results from the ranking exercises were used as prompts to further explore perspectives within the focus groups and interview discussions. Focus groups and interviews were audio-recorded and transcribed verbatim.

| Quantitative data analysis
Data from the ranking exercises were analysed using SPSS 21 to calculate the median and interquartile ranges (IQRs) for information content items, and mode of delivery options. The median, IQR and, if needed, the range were used to determine the order for both ranking exercises by participant group (VUE and PROSPECT). The mean age of the participants and the median length of the interview were also calculated for the two trial participant groups.

| Qualitative data analysis
The qualitative data generated were largely used to expand on the results of the ranking exercises. The transcribed manuscripts were compared to the original recordings to correct any discrepancies between the two. Data were analysed using the 'Framework' approach. 22,23 A priori themes were informed by the quantitative data on the most/least important items from the ranking exercise and any additional themes were generated inductively from the interview data. The first four transcripts were analysed using an open coding approach to develop the thematic framework. All coding themes initially identified in the data set by Jessica Wood were reviewed by Katie Gillies and Seonaidh C. Cotton for agreement. Full coding was then completed in NVivO-12, and data were charted onto a framework matrix. 24 Comparisons of data within and across the VUE and PROSPECT participant groups were performed and mapped. Quotes were then selected that helped to illustrate the overall groups' reasons for ranking items and to show any discordant views.

| Patient public involvement
This project involved contribution from public partners who were already working with a research team on a larger project exploring how to share trial results with participants (RECAP-researchregistry4085). Public partners contributed to focussed activities for this add-on Masters project. The ranking exercises and patient information leaflet were reviewed by two public representatives for ease of understanding and revised accordingly.

| Participant characteristics
Invitation packs were issued to 72 potential study participants from the PROSPECT trial; 10 agreed, and 2 focus groups were conducted. Fiftythree invitation packs were issued to potential study participants from the VUE trial, with six agreeing to be interviewed. All study participants were women, with a mean age of 71.6 years (SD = 9.7). The median focus group and interview times are shown in Table 1.

| General perspectives on information deemed important in trial results summaries
Before completing the ranking exercises, focus groups and interview participants were invited to suggest what information they deem to be important when receiving trial results. A number of participants suggested that information about implications for the future would be important, specifically in relation to what they could expect longer term.
These were split into some participants wanting to know the longerterm outcomes post trial treatment (i.e., long term follow-up), with others wanting to know how the trial would influence treatments offered within the NHS. Linked to this was women reporting that they wanted to know the overall 'success' of the trial, in her words what the

| Information content items deemed most important
Overall ranking summaries for the information content are presented in Table 2. There was discordance between the two groups in terms of information content items that were deemed the most important. The PROSPECT participants ranked 'If you can be given your individual results from the study' as the most important information content item, with the VUE participants ranking it as sixth. Some of the PROSPECT participants voiced poor outcomes following surgery as reasons for wanting to receive individual results and ranked this information as the most important as they felt that the overall 2-year results that they had received previously did not match their own experiences. In addition, some PROSPECT participants mentioned media attention as possibly contributing to wanting to receive their own individual results and why they ranked this information content item highly.
[When discussing the overall results in PROSPECT] 'Obviously we must be an unfortunate lot. I've spoken to different people that had it done, even people I was in the ward with, and there were five of us in at the same time,

| Preferred mode of delivery
The results from the ranking on the preference of how participants would want to receive results are presented in 'In-person' dissemination was ranked as the second most preferred method by the PROSPECT participants and third most preferred method by the VUE participants. When this was explored in further discussions, it was expressed by a number of the participants in both groups that there was interest in this method of dissemination, with many liking the idea of having someone to talk to and being in a group setting to be able to share their experiences and gain reassurance regarding surgical outcomes. The reasons given for not ranking 'in person' dissemination of results higher in some cases included a concern of others and one's own availability to come to a meeting.
Could it have been helpful, rather than having it in a newsletter, that you got the opportunity to talk to somebody-so either had a meeting like this where the researchers were here, and they could say, this is the study, and these were the results, and then it would give you the opportunity to have (to speak to others to know) …. you are not the only one it's happening to.

(PROSPECT Participant 4 focus group 2)
Yes, because I thought… it would be nice to see

| Receiving results
Both groups of participants were also asked if they could think of any situations in which they may not want to receive trial results. There was a strong overall consensus in both groups that participants should be provided with the results, with one participant then suggesting that the onus would then be on the participant to decide whether or not to read the results. Both participant groups ranked a thank you as the least important content item. It is possible that our study cohort felt that that they had already been thanked and acknowledged throughout their trial participation. A recent study found that participants felt that receiving results acknowledged their own individual contribution. 27 It is unclear if this was also the case in our study, although 'wanting to help other people' was identified as a key theme for the thank you item ranking the lowest. These findings align with previous research showing that altruistic reasons were often voiced by would-be trial participants as reasons for clinical trial participation. 28,29 Comparison of trial participant groups suggests that the experiences of trial participation may affect the information content that participants wish to know. Seeking reassurance from the trial results was a clear theme that contributed to the higher ranking of knowing which treatment was received and how to report side effects. In a previous survey of 3381 research participants, 70.6% wanted to have results that were specific to them. 15 Previous qualitative research has also shown that participants wish to receive information about the treatment that they received to better understand ongoing health issues and seek advice appropriately. 16,17 Differences between the PROSPECT and VUE participants' rankings may be attributable to the timing of our study in relation to the host trials and results received, that is, PROSPECT participants had already received the 2-year trial results, whilst VUE participants had not. However, these differences in whether participants had received trial result summaries or not were only considered post data analysis and as such not prospectively included for investigation. These differences in ranking may also be due to the difference in data collection methods for each cohort, that is, focus group versus individual interviews.
In both groups, the preferred mode of delivery of the trial results was by post, with reasons including having a hard copy to be able to refer back to and it being a more accessible mode of dissemination compared to those that required the use of and access to technology. These findings support previous work in this area, which found that participants wished to receive a hard copy of the results. 27,30 Having the results in an accessible format was a running subtheme and this finding may also relate to the older age of our study cohort. This was also highlighted in previous qualitative work that considered mode appropriateness according to the demographic, suggesting that older participants may not have access to certain modes of communication. 12,25 Although the HRA guidance does not currently suggest 'inperson' as a mode of delivery for dissemination, there appeared to be an interest in this option due to having someone to talk to and to share experiences with. 6 Previous work in this area reports similar findings. 31,32 In-person delivery of trial results may be geographically challenging in multicentre randomized-controlled trials. However, holding events online may overcome some of these challenges and could be an area explored by trial teams for future 'in-person' activities, but recognition of problems with digital access would also need to be considered and balanced. 33 Our findings suggested that participants may wish to receive the results in more than one format.
Previous research suggests that results could be made available to participants in a variety of ways rather than just providing one mode of delivery. 25 Working with patient partners to design the content and mode of delivery of trial results summaries will help to ensure that preferences are built into trial dissemination activities. | 425 evaluations to determine which methods are most effective could also add value, but determining the most appropriate outcomes to measure to determine whether this has been done well also warrants attention. All of these research activities should ensure that they include patients and the public as partners in the research process.

| STRENGTHS AND LIMITATIONS
A purposeful form of data sampling (rather than convenience sampling) may have provided a greater spread of opinions and views; however, we were limited to geographic location and to trial participants who had consented to be contacted about other related research. Another potential limitation of this study was the size of the sample (10 participants in two focus groups and 6 participant interviews). Participants were recruited from two trials in a similar disease area and patient population (gender and age); therefore, our findings may not be generalizable.
Demographic data were not collected, which may relate to participants' views and opinions expressed in the focus groups and interviews.
In addition, the preferred mode of dissemination could be related to the age group included in this study and could be a potential limitation of our study. A study by Purvis et al. 25

During the focus group
Start off by asking participants to talk about their experience of participating in PROSPECT.
• Do you remember receiving the results of the trial?
• What stood out for you from the results/what do you remember?
• Was there anything else you wanted to know; if yes, where from?
• Was the information presented in a way that was understandable?
-If yes what was helpful/if not, what was not clear?
• We sent the results by post in the form of a newsletter, how did you find this? Would you have liked a link to more information?
Would you have preferred it delivered differently, for example, face to face or the internet?

Interactive exercise
We are now going to complete an interactive exercise; some questions require you to rank items and others are yes/no questions.
Introduction to the ranking exercise: It is a bit like voting in the local elections when you need to rank your first choice of candidate, your second choice of candidate and so on. Rather than using a voting slip, we will be using some handheld devices to complete this. In VUE, they compared different operations for prolapse in women with or without a womb. There were two studies within VUE.
-Uterine trial (women with womb): Vaginal hysterectomy compared with an operation to suspend the uterus without removing it, and compared with suspending it via the abdomen (tummy).
For the purpose of today, we are not going to be discussing the VUE results, but I would like to find out about your experiences of taking part in a trial like VUE and your views on receiving results from trials like VUE.
• What do you remember being told when you consented to take part in VUE?
• If they do not mention results-ask them if they remember that.
• What kind of information would you like to see in a results sum- This study is all about receiving results -would you like to receive the results from this student project? Also, how would you like it delivered to you?
Finally, the information provided today will be used by the VUE team to better prepare the trial results summary for participants. This will be sent out in Spring next year.